Skip to content

GitLab

Unverified Commit 76d17c7e authored by lambert0312's avatar lambert0312 Committed by GitHub
Browse files

Fix shared experts fusion error without quantization (#5632)

parent 70d040f9
Hide whitespace changes
Inline Side-by-side
Showing with 1 addition and 1 deletion
python/sglang/srt/models/deepseek_v2.py
View file @ 76d17c7e
...@@ -1604,7 +1604,7 @@ class DeepseekV2ForCausalLM(nn.Module): ...@@ -1604,7 +1604,7 @@ class DeepseekV2ForCausalLM(nn.Module):
if self.n_share_experts_fusion > 0: if self.n_share_experts_fusion > 0:
weights_list = list(weights) weights_list = list(weights)
weights_dict = dict(weights_list) weights_dict = dict(weights_list)
if self.quant_config.get_name() == "w8a8_int8": if self.quant_config is None or self.quant_config.get_name() == "w8a8_int8":
suffix_list = [ suffix_list = [
"down_proj.weight", "down_proj.weight",
"down_proj.weight_scale", "down_proj.weight_scale",
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment